It wasn't a hack. One hidden bug took down Outlook and Teams in 74 countries. Here's what happened, and how to manage the risk of a vendor you can't replace.
The 43 minutes a month point is the one most small businesses haven't clocked. For a five-person office, the cheapest continuity is boring: a phone list of staff and key clients that doesn't live in Outlook, a backup channel everyone already knows to switch to, and someone who can update the website to say "email's down, please call". None of that needs a second vendor. Did you see many organisations with a fallback that actually worked on the day?
For small businesses like that, it's completely reasonable to prepare these few steps in advance and accept the rest of the risk. That's basically what risk management is all about.
I know organizations running on-prem Exchange servers that were nearly unaffected. And I know organizations that were ready for situations like this and simply followed a short protocol.
But from what I heard, most organizations had no idea what was going on and were struggling to keep their processes running.
The Manchester trial detail lands hardest for me. A court halting because email and chat went down shows how one vendor outage becomes a public service failure, not just an IT headache. Your point about the 99.9% SLA still allowing 43 mins of downtime is worth repeating to any board that thinks a contract equals cover.
Agreed, and that conversation has to happen before the outage, not after it. A named owner for the risk, with a date to report back, turns board attention into action.
Thank you for this post detailing steps on the recent outage. Some years ago I started disengaging with tech (including MS) because I did not appreciate what I saw as bully tactics in their businesses. It has been very time consuming. Also, these tech billionaires try to ensure there's no other replacement.
WhatsApp is not on my list of approved suppliers. Which means no contact with many people. But those critical to me were forced to engage with other suppliers to be able to contact me.
How fortunate was I when no outage was apparent! It will, I am sure, happen to me in the future…
Thank you for sharing this, and for reading so closely!
What you did is actually a great example of the article's main point: you reduced your dependence on a single provider.
And you're right, every provider will have its day eventually. Even the alternatives. The goal isn't to find a platform that never fails. It's making sure one failure doesn't stop everything.
The 43 minutes a month point is the one most small businesses haven't clocked. For a five-person office, the cheapest continuity is boring: a phone list of staff and key clients that doesn't live in Outlook, a backup channel everyone already knows to switch to, and someone who can update the website to say "email's down, please call". None of that needs a second vendor. Did you see many organisations with a fallback that actually worked on the day?
For small businesses like that, it's completely reasonable to prepare these few steps in advance and accept the rest of the risk. That's basically what risk management is all about.
I know organizations running on-prem Exchange servers that were nearly unaffected. And I know organizations that were ready for situations like this and simply followed a short protocol.
But from what I heard, most organizations had no idea what was going on and were struggling to keep their processes running.
The Manchester trial detail lands hardest for me. A court halting because email and chat went down shows how one vendor outage becomes a public service failure, not just an IT headache. Your point about the 99.9% SLA still allowing 43 mins of downtime is worth repeating to any board that thinks a contract equals cover.
Well, it showed pretty clearly who is willing to have these conversations when nothing is going on and who isn’t.
We should all realize that this isn’t something that IT can manage. These things require board-level executives to manage the risks.
Thank you for your comment!
Agreed, and that conversation has to happen before the outage, not after it. A named owner for the risk, with a date to report back, turns board attention into action.
Exactly. A named owner and a report-back date. That’s the key! :)
Thank you for your comments! Great to have these conversations!
Thank you for this post detailing steps on the recent outage. Some years ago I started disengaging with tech (including MS) because I did not appreciate what I saw as bully tactics in their businesses. It has been very time consuming. Also, these tech billionaires try to ensure there's no other replacement.
WhatsApp is not on my list of approved suppliers. Which means no contact with many people. But those critical to me were forced to engage with other suppliers to be able to contact me.
How fortunate was I when no outage was apparent! It will, I am sure, happen to me in the future…
Thank you for sharing this, and for reading so closely!
What you did is actually a great example of the article's main point: you reduced your dependence on a single provider.
And you're right, every provider will have its day eventually. Even the alternatives. The goal isn't to find a platform that never fails. It's making sure one failure doesn't stop everything.
As always you are absolutely correct! We have to find our own alternatives to inclusive technology.
Either find alternative or figure out what will happen when that SPOF fails. (And it will)